healthy diets asap – australian standardised …...healthy ‘discretionary’ food choices,...

14
STUDY PROTOCOL Open Access Healthy diets ASAP Australian Standardised Affordability and Pricing methods protocol Amanda J Lee 1* , Sarah Kane 2 , Meron Lewis 1 , Elizabeth Good 3 , Christina M Pollard 4,5 , Timothy J Landrigan 4 and Mathew Dick 3 Abstract Background: This paper describes the rationale, development and final protocol of the Healthy Diets Australian Standardised Affordability and Pricing (ASAP) method which aims to assess, compare and monitor the price, price differential and affordability of healthy (recommended) and current (unhealthy) diets in Australia. The protocol is consistent with the International Network for Food and Obesity / non-communicable Diseases Research, Monitoring and Action Supports (INFORMAS) optimal approach to monitor food price and affordability globally. Methods: The Healthy Diets ASAP protocol was developed based on literature review, drafting, piloting and revising, with key stakeholder consultation at all stages, including at a national forum. Discussion: The protocol was developed in five parts. Firstly, for the healthy (recommended) and current (unhealthy) diet pricing tools; secondly for calculation of median and low-income household incomes; thirdly for store location and sampling; fourthly for price data collection, and; finally for analysis and reporting. The Healthy Diets ASAP protocol constitutes a standardised approach to assess diet price and affordability to inform development of nutrition policy actions to reduce rates of diet-related chronic disease in Australia. It demonstrates application of the INFORMAS optimum food price and affordability methods at country level. Its wide application would enhance monitoring and utility of dietary price and affordability data from a health perspective in Australia. The protocol could be adapted in other countries to monitor the price, price differential and affordability of current and healthy diets. Keywords: Diet price, Food price, Diet affordability, Food affordability, Food policy, Food environments, Healthy diets, INFORMAS, Fiscal policy, Nutrition policy, Obesity prevention, Non-communicable disease, Monitoring and surveillance Background Poor diet is now the major preventable disease risk fac- tor contributing to burden of disease, globally and in Australia [1]. Less than 4 % of the population consume diets consistent with the evidence-based Australian Diet- ary Guidelines [2, 3]; on average, at least 35% of the total daily energy intake of adults and at least 39% of the energy intake of children [4] are now derived from un- healthy discretionaryfood choices, defined as foods and drinks high in saturated fat, added sugar, salt and/or alcohol that are not required for health [3]. Of particular concern is the contribution of poor diet to the rising rates of overweight and obesity. Based on measured height and weight, 25% of Australian children aged two to 17 years and 63% of Australian adults aged 18 years and over are now overweight or obese [5]. There is an urgent need for nutrition policy actions to help shift the current diet of the population towards healthy diets as recommended by the Australian Dietary Guidelines [3, 6]. The expense of healthy foods has been reported as a key barrier to consumption in Australia, particularly among low socioeconomic groups [711]. However, well-defined data in this area are lacking [6] as classifica- tion of healthyand unhealthyfoods and diets varies [12, 13] and the relative price of healthyand unhealthyfoods depends on the unit of measure (i.e. per energy unit, nutrient density, serve or weight) [14]. Compari- sons can be difficult particularly in the context of the total diet and habitual dietary patterns that are the major determinant of diet-related disease [3, 1517]. However, * Correspondence: [email protected] 1 The Australian Prevention Partnership Centre, The Sax Institute, 10 Jones Street, Ultimo, NSW 2007, Australia Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Lee et al. Nutrition Journal (2018) 17:88 https://doi.org/10.1186/s12937-018-0396-0

Upload: others

Post on 20-Aug-2020

3 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Healthy diets ASAP – Australian Standardised …...healthy ‘discretionary’ food choices, defined as foods and drinks high in saturated fat, added sugar, salt and/or alcohol that

STUDY PROTOCOL Open Access

Healthy diets ASAP – Australian StandardisedAffordability and Pricing methods protocolAmanda J Lee1* , Sarah Kane2, Meron Lewis1, Elizabeth Good3, Christina M Pollard4,5, Timothy J Landrigan4

and Mathew Dick3

Abstract

Background: This paper describes the rationale, development and final protocol of the Healthy Diets AustralianStandardised Affordability and Pricing (ASAP) method which aims to assess, compare and monitor the price, pricedifferential and affordability of healthy (recommended) and current (unhealthy) diets in Australia. The protocol isconsistent with the International Network for Food and Obesity / non-communicable Diseases Research, Monitoringand Action Support’s (INFORMAS) optimal approach to monitor food price and affordability globally.

Methods: The Healthy Diets ASAP protocol was developed based on literature review, drafting, piloting and revising,with key stakeholder consultation at all stages, including at a national forum.

Discussion: The protocol was developed in five parts. Firstly, for the healthy (recommended) and current (unhealthy)diet pricing tools; secondly for calculation of median and low-income household incomes; thirdly for store location andsampling; fourthly for price data collection, and; finally for analysis and reporting. The Healthy Diets ASAP protocolconstitutes a standardised approach to assess diet price and affordability to inform development of nutrition policyactions to reduce rates of diet-related chronic disease in Australia. It demonstrates application of the INFORMASoptimum food price and affordability methods at country level. Its wide application would enhance monitoring andutility of dietary price and affordability data from a health perspective in Australia. The protocol could be adapted inother countries to monitor the price, price differential and affordability of current and healthy diets.

Keywords: Diet price, Food price, Diet affordability, Food affordability, Food policy, Food environments, Healthy diets,INFORMAS, Fiscal policy, Nutrition policy, Obesity prevention, Non-communicable disease, Monitoring and surveillance

BackgroundPoor diet is now the major preventable disease risk fac-tor contributing to burden of disease, globally and inAustralia [1]. Less than 4 % of the population consumediets consistent with the evidence-based Australian Diet-ary Guidelines [2, 3]; on average, at least 35% of the totaldaily energy intake of adults and at least 39% of theenergy intake of children [4] are now derived from un-healthy ‘discretionary’ food choices, defined as foods anddrinks high in saturated fat, added sugar, salt and/oralcohol that are not required for health [3]. Of particularconcern is the contribution of poor diet to the rising ratesof overweight and obesity. Based on measured height and

weight, 25% of Australian children aged two to 17 yearsand 63% of Australian adults aged 18 years and over arenow overweight or obese [5]. There is an urgent need fornutrition policy actions to help shift the current diet ofthe population towards healthy diets as recommended bythe Australian Dietary Guidelines [3, 6].The expense of healthy foods has been reported as a

key barrier to consumption in Australia, particularlyamong low socioeconomic groups [7–11]. However,well-defined data in this area are lacking [6] as classifica-tion of ‘healthy’ and ‘unhealthy’ foods and diets varies[12, 13] and the relative price of ‘healthy’ and ‘unhealthy’foods depends on the unit of measure (i.e. per energyunit, nutrient density, serve or weight) [14]. Compari-sons can be difficult particularly in the context of thetotal diet and habitual dietary patterns that are the majordeterminant of diet-related disease [3, 15–17]. However,

* Correspondence: [email protected] Australian Prevention Partnership Centre, The Sax Institute, 10 JonesStreet, Ultimo, NSW 2007, AustraliaFull list of author information is available at the end of the article

© The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

Lee et al. Nutrition Journal (2018) 17:88 https://doi.org/10.1186/s12937-018-0396-0

Page 2: Healthy diets ASAP – Australian Standardised …...healthy ‘discretionary’ food choices, defined as foods and drinks high in saturated fat, added sugar, salt and/or alcohol that

the relative price and affordability of current and healthy(recommended) diets have been assessed rarely, as op-posed to the relative price of selected pairs of ‘healthy’and ‘less healthy’ foods [6].Various methods have been utilised to assess food

prices in Australia, such as Consumer Price Indexes(CPI) [18, 19] and supermarket price surveys, howeverthese usually tally the price of highly selected individualfood items and do not necessarily relate to relative costof the total habitual diet [6, 13]. A variety of ‘food bas-ket’ diet costing tools have also been developed at state,regional and community levels [13]. These methods havethe potential to measure the cost of a healthy diet. How-ever, dissimilarity of metrics is a recognised barrier tothe production of comparable data [7, 13].A recent systematic review of food pricing methods

used in Australia since 1995, identified 59 discrete sur-veys using five major food basket pricing tools (used inmultiple survey areas and multiple time periods) and sixminor food basket pricing tools (used in a single surveyarea or time period) [13]. No national survey had beenconducted. Survey methods differed in several metricsincluding: type and number of foods surveyed; applica-tion of availability and/or quality measures; definition ofreference households; calculation of household income;store sampling frameworks; data collection; and analysis.Hence results are not comparable across different loca-tions or different times [13]. With exception of Queens-land Health’s Healthy Food Access Basket tool revised in2015, [20] none of these fully align with a healthy diet asrecommended by the Australian Dietary Guidelines [3].Further, none accurately reflect current Australian diets[2, 4, 5, 13].Since 1995, the vast majority of ‘healthy’ food pri-

cing surveys in Australia have confirmed that: foodprices in rural and remote areas are up to 40% higherthan those in capital cities; lower socioeconomichouseholds need to spend a higher proportion of theirincome to procure healthy diets than other Austra-lians, and food prices generally increase over time [13,21]. Related calls for interventions, such as for freightsubsidies or food subsidies for low income groups inspecific regions have gone unheeded [22, 23]. Hence,it could be asserted that these surveys have hadlimited utility in informing fiscal and health policy[13]. As a result, there have been several calls for thedevelopment of standardised, healthy food and dietpricing survey methods nationally in Australia [24,25] and globally [6]. There is also a need forpolicy-relevant data [6, 26].The aim of relevant nutrition policy actions is to help

shift the current intake of the whole population to ahealthier diet consistent with dietary recommendations.Governments can manipulate food prices through a range

of complex policy approaches [6]. Three common strat-egies to increase the affordability of ‘healthy’ foods are:

� taxing ‘unhealthy foods’ (“fat taxes”) e.g. on sugarsweetened beverages;

� exempting ‘healthy foods’ from goods and servicetax (GST) or value added tax; and

� subsidising ‘healthy foods,’ such as through agriculturaland transport subsidies, retail price reductions, orvoucher systems targeted to vulnerable populationgroups [6].

Therefore, to inform relevant policy decisions, robustdata are required for both current (unhealthy) and healthy(recommended) diets [6]. With respect to food price andaffordability, the key health and nutrition policy relevantquestion to be answered by food pricing surveys is: “Whatis the relative price and affordability of ‘current’(unhealthy) and ‘healthy’ (recommended) diets?”While the potential effects of specific changes to fiscal

policy have been modelled [27, 28], recent ‘real life’ dataare lacking to inform policy decision making in Australia[29]. Assessment of the price, price differential andaffordability of a healthy diet (consistent with DietaryGuidelines) and current (unhealthy) diets (based onnational surveys), determined by standardised nationalmethods, would provide more robust data to informhealth and fiscal policy in Australia and monitor poten-tial fiscal policy interventions [13].There is a lack of such data globally; the current re-

search helps to address this, within the food price mod-ule of the International Network for Food and Obesity/non-communicable diseases Research, Monitoring andAction Support (INFORMAS) [6, 30]. Under the aus-pices of INFORMAS, the results of this study provide apotential globally-applicable stepwise food price and af-fordability monitoring framework that advocates ‘min-imal’, ‘expanded’ and ‘optimal’ approaches, to establishbenchmarks and monitor the cost of healthy food, mealsand diets; the level depends on availability of data andcountry capacity [6]. The novel INFORMAS ‘optimal’approach proposes concurrent application of two foodpricing tools to assess the price, price differential and af-fordability of a healthy diet (consistent with DietaryGuidelines) and current (unhealthy) diets (based on na-tional surveys). It requires assessment of household in-come, representative sampling and, ideally, stratificationby region and socio-economic status (SES).Based on the ‘optimal’ approach of the INFORMAS

diet price and affordability framework, we developed astandardised method to assess and compare the priceand affordability of healthy and current diets inAustralia, provide more robust, meaningful data to in-form health and fiscal policy in Australia, and develop

Lee et al. Nutrition Journal (2018) 17:88 Page 2 of 14

Page 3: Healthy diets ASAP – Australian Standardised …...healthy ‘discretionary’ food choices, defined as foods and drinks high in saturated fat, added sugar, salt and/or alcohol that

national data benchmarks with the potential for inter-national comparisons [29].This paper presents the resultant protocol for Healthy

Diets ASAP methods in Australia.

AimThe aim of this paper is to describe the developmentand final protocol of the Healthy Diets ASAP methods,based on the INFORMAS optimal price and affordabilityapproach. It details tools and methods to assist others toapply the approach in a standard manner, in order toenable comparison of the price, price differential andaffordability of healthy (recommended) and current(unhealthy) diets in Australia.

MethodsDevelopment of the healthy diets ASAP protocolBackground: Developing and piloting the initial diet pricingtools and methodsIn November 2013, all key Australian stakeholders gavein-principle support at a national teleconference for thedevelopment of national food price and affordabilitymonitoring methods based on the INFORMAS ‘optimal’approach. The development and pilot testing of themethods using readily available dietary data for fivehousehold structures in high socio-economic (SES) andlow SES areas is reported elsewhere [29]. The findingsconfirmed that the general approach could provide use-ful, meaningful data to inform potential fiscal and healthpolicy actions. Application of the diet pricing toolsaccurately reflected known composite food group ratios[2] and the proportion of the mean food budgetAustralian households spent on discretionary foods anddrinks in analysis of the Consumer Price Index (CPI)with respect to Australian Dietary Guidelines foodgroups [19]. However, internal validity testing suggestedthat construction of some of the initial diet pricing toolscould be improved to enhance accuracy [29]. For ex-ample, while performance of both diet pricing tools wasacceptable at household level, only the healthy diet pri-cing tool was acceptable at an individual level for alldemographics in the sample; the unhealthy (current) dietpricing tool could be improved for the 14 year old boyand both genders aged 70 years or over [29]. Further, po-tential systematic errors could be minimised by the utilisa-tion of detailed dietary survey data in the ConfidentialisedUnit Record Files (CURFs) from the Australian HealthSurvey (AHS) 2011–12 [31] and the Australian 2011–13food composition database [32], both of which were un-available at the time of the pilot study [29].Development of accepted, standardised diet pricing

methods also required agreement from all key stake-holders on the final approach, including accord on sys-tematic arbitrary decisions points around application of

the tools (such as whether to record the price of thenext largest or smallest packet if a particular size of foodwas unavailable in-store). There was also a desire to sim-plify methods to optimise uptake and utility.

Development and testing of diet pricing tools and processprotocolsThe final Healthy Diets ASAP protocols were developedin two phases.

Phase one: Revising and re-testing initial tools andmethods The food pricing tools were revised based onthe pilot outcomes [29] and feedback from internationalfood pricing experts (including at the Food Pricing Work-shop convened by authors AL and CP at the 14th Inter-national Society of Behavioural Nutrition and PhysicalActivity (ISBNPA) conference in Edinburgh May 2015).The revised unhealthy (current) diet pricing tools

reflected dietary data at the five-digit level by age andgender groupings [4] in the CURFs of the AHS 2011–12[31]. The most commonly available branded items andunit sizes in Australian supermarkets were identifiedfrom the pilot [29]. Other minor changes, and the rea-sons for these, are included in Table 1.The revised Healthy Diets ASAP diet pricing tools and

methods were applied to assess the price, price differen-tial and affordability of current and healthy diets in sixrandomly selected locations in two major cities (Sydney,New South Wales and Canberra, Australian Capital Ter-ritory) in November and December 2015. The prelimin-ary reports of these studies were provided to NSWHealth and ACT Health in early 2016. Colleagues inthese government departments provided feedback on therevised methods early March 2016.

Phase 2: Development of the final protocol At the na-tional Healthy Diets ASAP Methods Forum (the Forum)held in Brisbane on 10 March 2016, 25 expert stake-holders from academia, government jurisdictions andnon-government organisations (see Acknowledgements)worked together to finalise the Healthy Diets ASAP toolsand methods for national application in Australia.De-identified preliminary data from and feedback on thereports provided to NSW Health and ACT Health wereused to highlight methodological challenges and arbi-trary decision points during the Forum.Generally, the revised tools and methods applied in

Sydney and Canberra were confirmed at the Forum.However, some simplifications around arbitrary decisionpoints were recommended (Table 2).The revised tools and methods were finalised accord-

ing to the recommendations from the Forum. The re-sultant Healthy Diets ASAP protocol is described indetail in the results.

Lee et al. Nutrition Journal (2018) 17:88 Page 3 of 14

Page 4: Healthy diets ASAP – Australian Standardised …...healthy ‘discretionary’ food choices, defined as foods and drinks high in saturated fat, added sugar, salt and/or alcohol that

Following the Forum, the food price data collected inSydney and Canberra in late 2015 were reanalysed ac-cording to the Healthy Diets ASAP protocol and thepreliminary reports to NSW Health and ACT Healthwere finalised in May 2016.

ResultsThe healthy diets ASAP protocolThere are five parts to the Healthy Diets ASAP protocol.

The healthy diets ASAP protocol part one: Construct of thediet pricing toolsThere are two diet pricing survey tools: the current(unhealthy) diet pricing tool; and the healthy (recom-mended) diet pricing tool (Table 3). The diet pricing sur-vey tools include provision of quantities of food for areference household consisting of four people, includingan adult male 31–50 years old, an adult female 31–50 yearsold, a 14 year old boy and an 8 year old girl. An allowancefor edible portion/as cooked, as specified in AUSNUT2011–13 [32], is included in both diet pricing tools. Anypost plate wastage was not estimated or included.

The healthy diets ASAP current (unhealthy) diet pricingtool The current (unhealthy) diet pricing tool constitutesthe sum of the mean intake of specific foods and drinks,expressed in grams or millilitres, in each age/gendergroup corresponding to the four individuals comprisingthe reference household, as reported in the AHS 2011–

12 [31]. Foods are grouped according to stakeholder rec-ommendations (Table 2) and amounts consumed perday are derived from the CURFs at 5-digit code level[31]. The mean reported daily intake for each of the fourindividuals (Additional file 1) are multiplied by 14 andtallied to produce the quantities consumed per house-hold per fortnight. The amounts of foods and drinkscomprising the Healthy Diets ASAP current (unhealthy)diet for the reference household per fortnight is pre-sented in Table 3. The total energy content of the refer-ence household’s current diet is 33,860 kJ per day.Common brands of included food and drink items areincluded in the data collection sheet in Table 4.

The healthy diets ASAP healthy (recommended) dietpricing tool The healthy diet pricing tool reflects therecommended amounts and types of foods and drinksfor the reference household for a fortnight, consistentwith the Australian Guide to Healthy Eating and theAustralian Dietary Guidelines [3]. The amounts are cal-culated from the daily recommended number of servingsand relevant serve size of foods for the age/gender andphysical activity level (PAL) of 1.5 of the four individualscomprising the reference household in the omnivorousFoundation Diet models [33]. As the Foundation Dietswere developed for the smallest adults (or in the case ofchildren, the youngest) in each age/gender group, theamounts of foods were increased by 20% for the 8 year old

Table 1 Minor revisions to the initial diet pricing tools and methods

Improvement Aim/rational/comment

Added bottled water, olive oil, and relatively healthypre-made “convenience” foods, such as sandwiches and cookedchicken, to the healthy (recommended) diet pricing tools

To enhance comparability with the current (unhealthy) diet pricing tools,that include comparable, but less healthy, options

Further aggregated nutritionally similar products with similar utility inboth diet pricing tools (for example, ‘cabana’ and ‘bratwurst’ weregrouped with ‘sausages’)

To minimize the number of items to be priced in-store to reduce surveyburden and cost

Included the same food groupings in the healthy food component ofboth current and healthy diet pricing tools

To simplify data collection, comparison between current and healthydiets and interpretation of results

Adjusted the diet of the 8 year old girl (who was the oldest in her age/gender group) from the base Foundation Diets levels, according to theprescribed methods of Total Diet modelling to inform the 2013 revisionof the Australian Guide to Healthy Eating of the Australian DietaryGuidelines [33]

To ensure adequate energy content of the constructed healthy(recommended) diet of the 8 year old girl in the reference household

Adjusted median household income at Statistical Area Level 2 (SA2) levelby relevant wage price index; clarified that available data sets at SA2 levelprovide median gross (i.e. not disposable) household income

To incorporate the effect of inflation. Median household income atsub-national (area) level is readily available from published governmentsources, so has been used frequently in calculation of food affordabilityin Australia [13]. However, published median household income data atarea level reflects gross (total) income and has not been adjusted foressential expenditures such as taxation, to reflect disposable householdincome; results should be interpreted accordingly.

Included a third option for estimating median disposable householdincome at the national level, for use in future national diet price andaffordability surveys.

To enhance comparability with low (minimum) disposable incomehousehold income, that is also calculated at the national level.Median disposable household income is available only at national levelcurrently; however data may be available at state/territory level in thefuture.

Lee et al. Nutrition Journal (2018) 17:88 Page 4 of 14

Page 5: Healthy diets ASAP – Australian Standardised …...healthy ‘discretionary’ food choices, defined as foods and drinks high in saturated fat, added sugar, salt and/or alcohol that

Table 2 Arbitrary decisions made by key stakeholders at the national Healthy Diets ASAP Forum

Decision Point Forum decision- standard protocol Rationale/other comments

Household structure

1. Number of household structures for whichresults are reported? (5 different structures weredeveloped in the pilot study)

• Report and compare results for onehousehold structure only

• Simpler to interpret and communicate resultsfor only one (common) household structure.Less analysis, and therefore resources, requiredto access diet prices, therefore the protocol ismore likely to be used

2. Composition of household structure? • 2 adults and 2 children:-adult male 31–50 yrs. old-adult female 31–50 yrs. old-boy 14 yrs. old-girl 8 yrs. old• Publish quantities of food to be includedfor a range of individuals (age/gender),in addition to those to be included in theselected household structure

• Most commonly used household structures inAustralian studies are 6 and 4 personhouseholds

• Of these household structures, use 4 as it isclosest to the median Australian householdsize of 3 persons

• Those interested in reporting results for otherhousehold structures (e.g. single parent orpensioners) could perform additional analysispost data collection

Data collection

3. Which products should be included?a) Branded?b) Generic?c) Cheapest?d) Sales items?e) Bulk deals?f) How should any optional data collected beidentified on the data collection forms?

a) Include most common market share brandedproducts (Australia wide)

b) Include generic products only if brandeditems are not available (but exclude ALDIsupermarket which tend to stock genericproducts). However, consider supportingoptional inclusion of cheapest generic item,including the special/sale price (also applies toinclusion of ALDI)c) Don’t specifically seek to include cheapestitem. However, consider supporting optionalinclusion of cheapest item, including the special/sale price (also applies to inclusion of ALDI)d) Exclude sales items (as above)e) Exclude bulk deals (i.e. two for the price ofone deals)Consider adding tick box in end column of datacollection form to record if costing generic/special/sale price items as optional extras

a) Include the most popular items reported inthe Australian Health Survey (AHS) 2011–13 ascurrent diet

b) Inclusion of generic items has potential tobias, affect comparability and distort resultsover time- but could be included ifconsumption data continues to suggestincreasing intake.

c) Cheapest price could also be collected toanswer an optional additional question, butinclusion of cheapest price, including of salesor generic items, has potential to bias, affectcomparability and distort results over time.d) As abovee) As above. If optionally, collecting thecheapest price, could use multi buy price bydividing to obtain single pricef) May need to use multiple data collectionforms for each store or add additional datacollection column if collecting optional prices

4: Unhealthy (current) diet pricing tool

a) Adjust for known under-reporting in AHS2011–12?

No adjustment; report as ‘best case scenario’ There are no robust data on which to baseadjustment factor, so could introduce error.Analysis is not adjusted for any other reasons.

b) Confirm coding for five food group anddiscretionary foods?

• Tinned meat and vegetables- code as ½ vegand ½ meat

• Tinned fruit – code as fruit• Ham salad sandwich- (replace with chickensalad sandwich) and code as 1/3 bread, 1/3veg, 1/3 chicken meat

• Choc-chip Muesli bar – code as discretionary• Flavoured milk – code as non-discretionary(decision consistent with ABS classification)

• Processed meats (e.g. ham) – code asdiscretionary

• Water – include ½ reported water intake asbottled water (costed) and ½ as tap water(not costed)

• Decisions should be consistent with codingused by the ABS in the AHS 2011–12

• Revisit decisions reassessed when theAustralian Dietary Guidelines (ADGs) arereviewed (i.e. in 5 years’ time)

5. Healthy (recommended) diet pricing tool

Should any extra healthy foods be included?Such as more convenience options, bottledwater? Is the healthy diet unrealistic withoutinclusion of some discretionary foods or drinks,such as alcohol?

• Water – include ½ reported water intake asbottled water

• Convenience items- confirmed inclusion ofroasted chicken and sandwich– no furtherinclusions

Use the ADG Modelled Foundation diets basedon rationale that:−63% Australian adults are overweight/obese-There was no adjustment for underreportingin current diet

Lee et al. Nutrition Journal (2018) 17:88 Page 5 of 14

Page 6: Healthy diets ASAP – Australian Standardised …...healthy ‘discretionary’ food choices, defined as foods and drinks high in saturated fat, added sugar, salt and/or alcohol that

girl who is the oldest in her height/age group, accordingto the recommendations [3].To ensure the most commonly consumed healthy foods

in Australia are used, food categories in the healthy dietpricing tool are the same as those in the current diet pri-cing tool (but differ in quantity). A variety of fresh, canned,frozen and dried foods is included. For example, represen-tative categories of fresh produce reflect common fruit andvegetables available all year round in Australia. Luxuryproducts, such as imported fruit and vegetables (particu-larly those out of season) and foods with very high cost perkilogram (e.g. oysters, smoked salmon) are excluded. Some‘convenience’ foods are included in the healthy diet pricingtool as per stakeholder decisions (Table 2).Consistent with Australian recommendations [3], the

healthy diet pricing tool does not contain any discretion-ary choices. It includes: grain (cereal) foods, in the ratio66% wholegrain and 33% refined varieties; cheese, milk,yoghurt and calcium-fortified plant based alternatives,

mostly (i.e. > 50%) reduced fat, with a maximum of 2–3serves of high fat dairy foods (cheese) per person perweek; lean meat (beef, lamb, veal, pork), poultry andplant-based alternatives (with no more than 455 g redmeat per person per week); a minimum of 140 g and upto 280 g fish per person per week; up to 7 eggs per personper week; a selection of different colours and varieties ofvegetables (green and brassica, orange, legumes, starchyvegetables, other vegetables) with a minimum 350 g perday for adults; a variety of fruit with a minimum of 300 gper day for adults; and an allowance of unsaturated oils orspreads or the nuts/seeds from which they are derived[33]. The daily quantities of food categories recommendedfor each individual (age/gender) in the reference house-hold (Additional file 2) are multiplied by 14 and tallied toprovide quantities per fortnight (Table 3).The amounts of foods and drinks comprising the

Healthy Diets ASAP healthy (recommended) diet for thereference household per fortnight are presented in Table 3.

Table 2 Arbitrary decisions made by key stakeholders at the national Healthy Diets ASAP Forum (Continued)

Decision Point Forum decision- standard protocol Rationale/other comments

• Alcohol – do not include -Most Australians are not expending enoughenergy to allow for additional energy intakefrom any discretionary foods or drinks- The healthy diet should be aspirational,and reflect that associated with optimalhealth outcomes

6: Income data

Should mean or median income be used?What assumptions should be used todetermine indicative low income?

• Include both median household (HH)income from published data and calculatedlow (minimum) disposable income household(HH) income (confirmed assumptions used inpilot calculations)

• Also consider reporting results against theAustralian poverty line

• Median HH income is specific to location,but is pre-tax i.e. not disposable income

• Low income HH calculation is not specificto location apart from rent (which is setlow so rarely changes)

• Poverty line is lower than 50% of theAustralian median HH income

• Median household income and indicativelow (minimum)vs disposable household income are notcomparable

7: Sampling framework

Sampling frameworks: which areas, stores,distances (e.g. 7 km radius of centre of SA2area) should be included?

• Sampling approach SA2 stratified by Indexof Relative Socio-Economic Disadvantage forAreas (SEIFA) and including all stores within aspecific radius confirmed (ALDI excluded ininitial methods as above)

• Requested further work to determinecalculating distance away from centre forinclusion of stores

• Methods of randomisation trialled isappropriate and feasible

• 7 km radius of inclusion may not beappropriate for all locations, particularlyin rural areas

8: Data collection protocols

Prioritisation of sizes and branding for pricing,as proposed on data collection sheet

• Proposed detailed methods confirmed e.g. sizeprescribed but if not available take next largersize first

• Detailed methods proposed are appropriate• Reflect common current practice in mostlocations; clear and concise; easy to follow

9: Definition of affordability

Should affordability level be set at 25% or 30%of disposable HH income?

• May need to assess both (post hoc) but initiallyuse 30% pending further review of the literatureand international consultation

• Based on most commonly used definitionin international literature from high incomecountries

Lee et al. Nutrition Journal (2018) 17:88 Page 6 of 14

Page 7: Healthy diets ASAP – Australian Standardised …...healthy ‘discretionary’ food choices, defined as foods and drinks high in saturated fat, added sugar, salt and/or alcohol that

Table 3 Composition of the current (unhealthy) and healthy(recommended) diets for the reference householda per fortnight

Food or drink Quantity

Current (unhealthy) diet

Bottled water, still (ml) 5296

Artificially sweetened ‘diet’ soft drink 2391

Fruit

Apples, red, loose (g) 3497

Bananas, Cavendish, loose (g) 899

Oranges, loose (g) 1664

Fruit salad, canned in juice (g) 2046

Fruit juice 3026

Vegetables

Potato, white, loose (g) 1460

Sweetcorn, canned, no added salt (g) 206

Broccoli, loose (g) 422

White cabbage, loose (g) 235

Iceberg lettuce, whole (g) 795

Carrot, loose (g) 753

Pumpkin (g) 240

Four bean mix, canned (g) 74

Diced tomatoes, canned, in tomato juice(g) 234

Onion, brown, loose (g) 84

Tomatoes, loose (g) 488

Frozen mixed vegetables, pre-packaged (g) 1184

Frozen peas, pre-packaged (g) 273

Baked beans, canned (g) 369

Salad vegs in sandwich 120

Veg in tinned meat and vegetable casserole (g) 646

Grain (cereal) foods

Wholegrain cereal biscuits Weet-bix™ (g) 430

Wholemeal bread, pre-packaged (g) 1054

Rolled oats, whole (g) 870

White bread, pre-packaged (g) 3033

Cornflakes (g) 680

White pasta, spaghetti (g) 1326

White rice, medium grain (g) 1622

Dry water cracker biscuit (g) 258

Bread in sandwich 120

Meats, poultry, fish, eggs, nuts and seeds

Beef mince, lean (g) 267

Lamb loin chops (g) 257

Beef rump steak (g) 1056

Tuna, canned in vegetable oil (g) 1052

Whole barbeque chicken, cooked (g) 1661

Eggs (g) 872

Table 3 Composition of the current (unhealthy) and healthy(recommended) diets for the reference householda per fortnight(Continued)

Food or drink Quantity

Meat in tinned meat and vegetable casserole (g) 646

Chicken in sandwiches 120

Milk, yoghurt, cheese and alternatives

Cheddar cheese, full fat (g) 624

Cheddar cheese, reduced fat (g) 44

Milk, full fat (ml) 5961

Milk, reduced fat (ml) 2929

Yoghurt, full fat plain (g) 204

Yoghurt, reduced fat, flavoured (vanilla) (g) 676

Flavoured milk (ml) 2416

Canola margarine (g) 170

Sunflower oil (ml) 7

Olive oil (ml) 7

Discretionary choices

Beer, full strength (ml) 4661

White wine, sparkling (ml) 863

Whisky (ml) 266

Red wine (ml) 1078

Butter (g) 280

Muffin, commercial (g) 1455

Cream-filled sweet biscuit, pre-packaged (g) 496

Muesli bar, pre-packaged (g) 373

Peanuts, salted (g) 255

Pizza, commercial (g) 1182

Savoury flavoured biscuits (g) 222

Confectionary (g) 418

Chocolate (g) 441

Sugar sweetened beverages (Coca Cola) (ml) 12,012

Meat pie, commercial (g) 1638

Frozen lasagne, pre-packaged (g) 4322

Hamburger, commercial (g) 2413

Beef sausages (g) 1048

Ham (g) 189

Potato crisps, pre-packaged (g) 518

Potato chips, hot, commercial (g) 670

Ice cream (g) 1830

White sugar (g) 564

Salad dressing (ml) 277

Tomato sauce (ml) 569

Chicken soup, canned (g) 1340

Orange juice (ml) 3027

Fish fillet crumbed, pre-packaged (g) 302

Lee et al. Nutrition Journal (2018) 17:88 Page 7 of 14

Page 8: Healthy diets ASAP – Australian Standardised …...healthy ‘discretionary’ food choices, defined as foods and drinks high in saturated fat, added sugar, salt and/or alcohol that

The total energy content of the household’s healthy diet is33,610 kJ per day. Common brands of included food anddrink items are included in the data collection sheet inTable 4.

Diet pricing tools for additional household structuresSeveral stakeholders requested (Table 2) that the compos-ition of current (unhealthy) and healthy (recommended)diets be provided for four other household compositionscommonly investigated in Australia1 (for example, for sin-gle parent or pensioner households) so that additionaldata analysis could be performed. These data are includedin Additional file 3.

Validity of the diet pricing survey tools Convergentvalidity of the constructed healthy and current diet pri-cing survey tools for each age/gender group was assessedby energy and macronutrient analysis using FoodWorks7 Professional [34] computer program installed withAUSNUT 2011–13 [32] (the food composition databaseused to analyse the AHS) and comparing results withAustralian Nutrient Reference Values [35] and nutrientresults from the AHS 2011–12 respectively [5, 31]. Theresults are presented in Additional file 4. As deemed ac-ceptable for modelling outputs to develop the AustralianGuide to Healthy Eating, [33] the energy content of theconstructed healthy diet pricing tool is within 5% of theFoundation Diet levels and the macronutrient profilesare within the recommended ranges for more than 97%of values for all age/gender groups. Similarly, the energycontent of the current diet pricing tool is within 5% ofthe reported energy intakes of the AHS 2011–2012 [4]for all individuals.Internal validity indicators, such as the ratio of fruit

and vegetables content between the healthy and currentdiet pricing tools (approximately 2:1) are consistent with

Table 3 Composition of the current (unhealthy) and healthy(recommended) diets for the reference householda per fortnight(Continued)

Food or drink Quantity

Instant noodles, wheat based (g) 381

Healthy (recommended) diet

Bottled water, still (ml) 5296

Fruit

Apples, red, loose (g) 5460

Bananas, Cavendish, loose (g) 5460

Oranges, loose (g) 5460

Vegetables

Potato, white, loose (g) 2320

Sweetcorn, canned, no added salt (g) 1160

Broccoli, loose (g) 1470

White cabbage, loose (g) 1470

Iceberg lettuce, whole (g) 1470

Carrot, loose (g) 2205

Pumpkin (g) 2205

Four bean mix, canned (g) 1005

Diced tomatoes, canned, in tomato juice(g) 1638

Onion, brown, loose (g) 1638

Tomatoes, loose (g) 1638

Frozen mixed vegetables, pre-packaged (g) 1638

Frozen peas, pre-packaged (g) 1638

Baked beans, canned (g) 1005

Salad vegs in sandwich 120

Grain (cereal) foods

Wholegrain cereal biscuits Weet-bix™ (g) 2216

Wholemeal bread, pre-packaged (g) 4272

Rolled oats, whole (g) 6648

White bread, pre-packaged (g) 893

Cornflakes (g) 670

White pasta, spaghetti (g) 2042

White rice, medium grain (g) 2042

Dry water cracker biscuit (g) 781

Bread in sandwich 120

Meats, poultry, fish, eggs, nuts and seeds

Beef mince, lean (g) 1168

Lamb loin chops (g) 1169

Beef rump steak (g) 1172

Tuna, canned in vegetable oil (g) 1841

Whole barbeque chicken, cooked (g) 1471

Eggs (g) 2208

Peanuts, roasted, unsalted (g) 780

Chicken in sandwiches 120

Table 3 Composition of the current (unhealthy) and healthy(recommended) diets for the reference householda per fortnight(Continued)

Food or drink Quantity

Milk, yoghurt, cheese and alternatives

Cheddar cheese, full fat (g) 704

Cheddar cheese, reduced fat (g) 516

Milk, full cream (ml) 6438

Milk, reduced fat (ml) 12,000

Yoghurt, full fat plain (g) 2576

Yoghurt, reduced fat, flavoured (vanilla) (g) 5100

Canola margarine (g) 412

Sunflower oil (ml) 291

Olive oil (ml) 291aThe reference household comprises four people: adult male 19–50 yrs. old;adult female 19–50 yrs. old; boy 14 yrs. old; girl 8 yrs. old

Lee et al. Nutrition Journal (2018) 17:88 Page 8 of 14

Page 9: Healthy diets ASAP – Australian Standardised …...healthy ‘discretionary’ food choices, defined as foods and drinks high in saturated fat, added sugar, salt and/or alcohol that

available published data [2, 31] and recommendations[33]. Further, the proportion of household food expend-iture on discretionary items (around 58%) [29] is similarto that described by the ABS (58.2%) using differentmethods based on household expenditure [19]. Hencethe tools appear valid for use in estimating the cost ofcurrent and healthy diets.

The healthy diets ASAP protocol part two: Location andstore sample selectionA random sample of the Statistical Area Level 2 (SA2)locations in each town is selected to achieve a represen-tative sample. SA2 locations are stratified by the Indexof Relative Socio-Economic Disadvantage for Areas(SEIFA) quintile using information and maps availableon the ABS website [36–38] Following sample size cal-culations, the required number of SA2 locations withinSEIFA Quintile 1, 3 and 5 are selected randomly forparticipation. Food outlets within seven kilometres bycar of the centre of each SA2 location are identified withGoogle™ Maps [39] and included in the surveys. Storesto survey include one outlet of all supermarket chains(in trials these were Coles™, Woolworths™ and Independ-ent Grocers Australia (IGA™), Supabarn™ and ALDI™),‘fast-food’/take-away outlets (a Big Mac™ hamburgerfrom the McDonald’s™ chain; pizza from the Pizza Hut™chain; fish and chips from independent outlets) and twoalcoholic liquor outlets closest to the geographical centreof each SA2 location.

The healthy diets ASAP protocol part three: Collecting andentering food price dataThe Healthy Diets ASAP diet price survey data collec-tion form (Table 4) combines the items included in thecurrent diet and the healthy diet for convenience andutility. The agreed price data collection protocol is pre-sented in Table 5 and is printed on each data collectionform. Research assistants are trained to use the formand follow the price collection protocol strictly. Pricesare collected within the same 4 week/monthly period, asprices change over time.Permission to participate is sought from each store

manager prior to data collection.Data entry and analysis sheets have been developed

using Excel™ spreadsheets [40]. Double data entry is rec-ommended to minimise error. Data are cleaned andchecked. Any missing values are imputed to ascribe themean price of the same food item in all other relevantoutlets in the same SA2 area. Data analysis tools areavailable from the corresponding author. As has beenachieved previously for the Victorian Health Food Ac-cess Basket [41], the Healthy Diets ASAP App is underdevelopment to streamline data collection and analysisand reduce error.

Table 4 Healthy Diets ASAP (Australian Standardised Affordabilityand Price) Survey Data Collection Form

Lee et al. Nutrition Journal (2018) 17:88 Page 9 of 14

Page 10: Healthy diets ASAP – Australian Standardised …...healthy ‘discretionary’ food choices, defined as foods and drinks high in saturated fat, added sugar, salt and/or alcohol that

The healthy diets ASAP protocol part four: Determination ofhousehold incomeHousehold income is determined by either of threemethods, depending on the purpose of the study and thegranularity of available data.

Median household gross income at area level InAustralia, national census data is the only source of SA2level household income data and is provided only attotal (gross) level. Median gross household income is de-termined per week (before taxation, rent and other ex-penses) in each SA2 area by entering relevant post codesinto the Community Profile data calculator [42] that isbased on the 2011 Census results [36], adjusted for thewage price index (for example, there was an increase of11.1% from September 2011 to September 2015) andmultiplying by two to derive median household incomein each SA2 area per fortnight. Details and examples areprovided in Additional file 5.

Indicative low (minimum) disposable householdincome Indicative low (minimum) income of the refer-ence household (and other households of interest to spe-cific stakeholders) is calculated based on the level ofminimum wages [43] and determination of the welfarepayments provided by the Department of Human Ser-vices [44] as per the methods used by the QueenslandDepartment of Health [20]. Assumptions are made foremployment, housing type, disability status, savings and

investments, child support, education attendance andimmunisation status of children (Table 6). As welfarepolicy actions can change, the most recent schedulesshould be used. Where it is higher than the minimumthreshold, the indicative low (minimum) household in-come is adjusted for taxation payable [45] so also repre-sents minimum household disposable income. Detailsand examples are provided in Additional file 6.

Median household disposable income at nationallevel For assessment of diet affordability at the nationallevel, median equivalised disposable household incomefor the reference family composition is sourced from theSurvey of Income and Housing [46].

The healthy diets ASAP protocol part five: Data analysisand reportingThe price of the healthy (recommended) and current di-ets in each store and the mean price for each SEIFAquintile is calculated for the reference household com-position in each of areas surveyed in each city. Resultscan be presented in a range of metrics, including thecost of the total diets per household per fortnight, andthe cost of purchasing specific five food group and dis-cretionary foods and drinks (including policy relevantitems such as alcohol, ‘take-away foods’ and sugar-sweet-ened beverages). The results for the current (unhealthy)diet and healthy (recommended) diet are compared todetermine the differential.Affordability of the healthy and current diets for the

reference household is determined by comparing thecost of each diet with the median gross householdincome (Additional file 5) and also with the indicativelow (minimum) disposable income of low income house-holds (Additional file 6). Where a representative nationalsurvey of diet prices has been conducted, affordability ofthe healthy and current diets for the reference householdis determined by comparing the cost of each diet withthe median equivalised disposable income [46] and withthe indicative low (minimum) disposable income of lowincome households. Internationally, a benchmark of 30%

Table 5 Healthy Diets ASAP food price data collection protocol

1. Record the usual price of an item, i.e. do not collect the sale/specialprice unless it is the only price available (if so, note in comment column)2. Look for the specified brand and specified size for each food item,and record the price• If the specified brand is not available: Choose the cheapest brand(non-generic) available in the specified size. Note this brand in the“Your brand” column• If the specified size is not available: Choose the nearest larger sizein the specified brand. If a larger size is not available, choose thenearest smaller size. Note this size in the “Your size” column• If both the specified brand and specified size are not available:Choose the cheapest in the nearest larger size of another brand(non-generic). If a larger size is not available, choose the nearestsmaller size• If multiple brands are specified, record the price of the cheapest oneand note brand in the “Your brand” column• If the item is only available in a generic form (e.g. Home Brand,Coles, Woolworths Select, Black and Gold) choose the most expensivegeneric item in the specified size. If the specified size is not available,choose the nearest larger size. If a larger size is not available, choosethe nearest smaller size. Note the generic name in the “Your brand”and the size in the “Your size” columns

3. Loose produce: choose the usual cheapest price per kg of the varietynot on special. If the only variety available is on special, record thespecial price and note in comments column4. Peanuts: choose the branded packet size closest to 250 g. Ifpackaged, roasted, unsalted peanuts are not available, record the priceof the loose ‘bulk scoop & weigh’ roasted, unsalted peanuts per 100 g5. Check all data are collected and recorded as above, before leaving store

Table 6 Assumptions applied to determine the indicative low(minimum) disposable household income of the reference household

The reference household consists of an adult male, an adult female, a14 year old boy and an 8 year old girl• The adult male works on a permanent basis at the national minimumwage ($17.29 per hour)for 38 h a week

• The adult female works on a part-time basis at the nationalminimumwage ($17.29 per hour)for 6 h a week

• Both children attend school and are fully immunised• None of the family are disabled• The family has some emergency savings that earn negligible interest

Lee et al. Nutrition Journal (2018) 17:88 Page 10 of 14

Page 11: Healthy diets ASAP – Australian Standardised …...healthy ‘discretionary’ food choices, defined as foods and drinks high in saturated fat, added sugar, salt and/or alcohol that

of income has been used to indicate affordability of adiet [6, 10].Data files can be manipulated to investigate the effects

of potential fiscal policy changes on the affordability ofcurrent (unhealthy) and healthy (recommended) dietsfor the reference household. The price of the relevantfoods and drinks can be modified readily to highlight thelikely ‘real-world’ impacts of different scenarios, for ex-ample, to investigate the potential extension of theGoods and Services Tax (GST) on basic healthy foods[47], or the potential application of different levels oftaxation on sugary drinks in Australia [29].

DiscussionThere are several methodological limitations inherent inthe Healthy Diets ASAP protocols. Given that it is basedon national reported mean dietary intakes, the cost ofthe current (unhealthy) diet is unlikely to be the same asactual expenditure on food and drinks in specific areasand among specific groups [48]. Other assumptionscommonly made in similar apparent consumption andhousehold expenditure surveys include that food isshared equitably throughout the household, that there isno home food production and minimal wastage. Nutri-tionally similar products were aggregated to minimise thenumber of items included in the diet pricing tools, butproducts were not necessarily homogenous in terms ofprice. However, similar healthy food items were includedin each diet to try to minimise any unintended effects.Ideally, the specific foods included in both diet pricing

tools are culturally acceptable, commonly consumed,widely available, accessible and considered ‘every day’ ra-ther than luxury items. As the foods and drinks includedin the current diet pricing survey tool reflect actual con-sumption data, it was presumed that they were deemedby the population as a whole as meeting these require-ments. No adjustments were made for costs such astransport, time, cooking equipment and utilities; as theseapply to both current and healthy diets, assessment ofthe price differential between the two can help controlfor some of these hidden costs to some extent. However,these hidden costs would increase actual diet costs anddecrease affordability of the diets.No adjustments were made to account for the marked

under-reporting in the AHS 2011–12 [4], reported dietaryvariability amongst different groups other than age/genderstratification, or the greater proportion of pre-prepared‘convenience’ items in the current diet pricing tool com-pared with the healthy diet pricing tool. Given the highrates of overweight/obesity in Australia, the FoundationDiets were prescribed for the shortest and least active ineach age group according to the modelling that informedthe Australian Guide to Healthy Eating [33]; however this

would under-estimate the requirements of taller, more ac-tive and healthy weight individuals.No attempt was made to control the price of the healthy

diet pricing tool or the current diet pricing tool for energy,as the diets are constructed on recommended energylevels and actual reported levels of energy respectively.Further, the energy content of each tool is a determinantvariable that directly affects diet-related health outcomes[18, 49]. As most Australians are already overweight orobese, increasing recommended energy requirements inexcess of Foundation Diets is not consistent withoptimum health outcomes [33]. As the key exposure vari-able affecting the life time risk of diet-related disease isthe total diet and dietary patterns, approaches such as thisthat compare metrics of actual current diets with recom-mended diets are more pertinent to the health policy de-bate than the more common, but limited, studies into therelative price of selected ‘healthy’ and ‘unhealthy’ foods orsingle ‘optimised’ diets [18, 50, 51].While a benchmark of 30% of income has been used to

indicate affordability of diet internationally and inAustralia [6, 9, 10] it is not clear from the literaturewhether this income comparator is gross income or dis-posable income [6]. Using disposable income to estimateaffordability better reflects the capacity of a household toafford food/diets [52, 53]; using gross income is a moreconservative approach as it does not take taxation into ac-count. However, in Australia currently, median disposablehousehold income data are readily available only at na-tional level [46]; at area level only median gross householdincome data are readily available. Further, the compositionof the reference household does not align necessarily withthat of households in the census in all areas. Comparingdiet price with indicative low (minimum) disposablehousehold more accurately estimates affordability of dietsin vulnerable groups. However, the tax paid component ofindicative low (minimum) disposable household incomecan be removed to improve comparability with estimatesof affordability determined by application of gross medianhousehold income.Arbitrary decision points occur around sampling frame-

works, data collection protocols (for example, selection ofcheapest comparable generic item if the branded item isunavailable in any size), analysis and presentation of re-sults, data sources and definitions of family and householdincome and composition. Such methodological limitationsare common to other food price studies. In order for finalmethods to be replicable, agreement among key stake-holders including end users on each of these decisionpoints at the Healthy Diets ASAP Forum was invaluable.Publication of detailed protocols is essential to support up-take, replicability, fidelity and transparency of the method.The detailed dietary survey data required to produce the

current (unhealthy) diet pricing tool and the modelling

Lee et al. Nutrition Journal (2018) 17:88 Page 11 of 14

Page 12: Healthy diets ASAP – Australian Standardised …...healthy ‘discretionary’ food choices, defined as foods and drinks high in saturated fat, added sugar, salt and/or alcohol that

data required to produce the healthy (recommended) dietpricing tool are not easily accessible in all countries andtechnical capacity to analyse individual records may belimited. Therefore, this optimal approach may be toocomplex for application to assess and monitor the price ofdiets from a health perspective globally. However, there ispotential for the diet pricing tools to be adapted for use inother countries by substitution of food components withcommonly-consumed local equivalents, dietary analysisand testing.

ConclusionThe development of standardised Healthy Diets ASAPmethod protocols provides an example of how theINFORMAS optimal food price and affordability methodscan be adapted at country level to help develop standar-dised, policy relevant diet price assessment, monitoringtools and benchmarks. The approach can be used to as-sess the price, price differential and relative affordability ofcurrent (unhealthy) and healthy (recommended) diets andinform scenario modelling of potential fiscal and nutritionpolicy actions.The Healthy Diets ASAP method satisfies long-standing

calls for the development of a nationally standardisedapproach to assess food prices from a health perspective,supporting comparison of results from different locationsand over time, in Australia.The protocol could be adapted in other countries to

benchmark and monitor the price, price differential andaffordability of current and healthy diets globally.

Endnotes1Additional data are provided for:

� Household 1(n = 6): adult male 31–50 yrs. old; adultfemale 31–50 yrs. old; older female 70+ yrs. old; boy14 yrs. old; girl 8 yrs. old; boy 4 yrs. old

� Household 2 (n = 3): single parent with 2 children:adult female 31–50 yrs. old; boy 14 yrs. old; girl8 yrs. old

� Household 3 (n = 1): single unemployed person:adult male 31–50 yrs. old

� Household 4 (n = 2): older couple with no children:senior adult male 70+ yrs. old; senior adult female70+ yrs. old: pensioners

Additional files

Additional file 1: Current (unhealthy) Diets: Mean daily intake ofrepresentative categories of foods and drinks for individuals (age/gender)comprising the reference household, and other common households.(DOCX 43 kb)

Additional file 2: A. Foundation diet recommended serves of foods perweek for individuals (NHMRC 2011) comprising the reference household

and other common households. B. Healthy (recommended) Diets:Recommended serves per day of food groups and amounts ofcomposite foods and drinks for individuals comprising the referencehousehold, consistent with Foundation Diets (NHMRC 2011) includingcommonly-consumed brands. (DOCX 55 kb)

Additional file 3: Composition of the current (unhealthy) diet andhealthy (recommended) diet for four additional households (HH1, HH2,HH3, HH4)1 per fortnight. (DOC 188 kb)

Additional file 4: Energy and nutrient analysis of individual current andhealthy diet baskets compared to results of the AHS and Foundation Dietmodelling. (DOCX 36 kb)

Additional file 5: Median income determination by SA2 Example-Median income data from the 2011 Census, ABS Community Profiles ofSA2 areas for six SA2 locations in Sydney, NSW*. (DOCX 36 kb)

Additional file 6: Calculations of low (minimum) disposable householdincome data from welfare data – Example. (DOCX 34 kb)

AbbreviationsASAP: Australian Standardised Affordability and Pricing methods;CURFs: Confidential Unit Record Files

AcknowledgementsThank you to Research Assistants Dr. Janice Lee and Ms. Sandra Mitchell whocollected food prices in two major Australian cities in November and December2015. Thank you to all those academic, public sector and non-governmentorganisation representatives who attended the National Healthy Diets ASAPForum, Brisbane, March 2016 and contributed to decision-making to informdevelop the final Healthy Diets ASAP methods, including Ms. Kathy Chapman,Prof Karen Charlton, Ms. Megan Cobcroft, Ms. Megan Ferguson, Prof DanielleGallegos, Ms. Leah Galvin, Ms. Holly Jones, Dr. Carolyn Keogh, Dr. Sally McKay,Mr. Leigh Merrington, Ms. Sandra Murray, Ms. Lesley Paton, Dr. Claire Palermo,Dr. Melanie Pescud, Ms. Emma Slaytor, Ms. Beth Thomas, Ms. Carrie Turner,Ms. Julie-Anne McWhinnie together with all authors. Thank you to INFORMASmembers.

FundingFinancial support to develop the national Healthy Diets ASAP methods wasprovided by The Australian Prevention Partnership Centre through the NHMRCpartnership centre grant scheme (Grant ID: GNT9100001) with the AustralianGovernment Department of Health, NSW Ministry of Health, ACT Health, HCF,and the HCF Research Foundation. The funders had no role in the design,analysis or writing of this manuscript.

Availability of data and materialsThe datasets supporting the conclusions of this article are included withinthe article and its additional files.

Authors’ contributionsAL led the project, developed concepts, constructed the current diet pricingtool, finalised the healthy diet pricing tool, developed sampling methods,convened and chaired the national stakeholders forum and drafted themanuscript; SK assisted in constructing the current diet pricing tool andfinalising the healthy diet pricing tool, transposed, cleaned and analyseddata and assisted with the national stakeholders forum; ML assisted withtransposing and analysing food price data, finalised the household incomeassessment protocol and assisted with the national stakeholders forum; EGdeveloped an early draft of the household income assessment protocol; CPprovided conceptual advice; TL accessed and analysed dietary intake datafrom the Confidential Unit Record Files (CURFs) of the Australian HealthSurvey 2011–13 (ABS 2013a) to inform development of the current dietpricing tool and advised on methods to determine household income; MDdeveloped an early draft of the healthy diet pricing tool. All co-authorsreviewed drafts of the paper and contributed to the final manuscript.

Ethics approval and consent to participateThe QUT University Human Research Ethics Committee assessed this studyas meeting the conditions for exemption from Human Research EthicsCommittee review and approval in accordance with section 5.1.22 of theNational Statement on Ethical Conduct in Human Research (2007); the

Lee et al. Nutrition Journal (2018) 17:88 Page 12 of 14

Page 13: Healthy diets ASAP – Australian Standardised …...healthy ‘discretionary’ food choices, defined as foods and drinks high in saturated fat, added sugar, salt and/or alcohol that

exemption number is 1500000161. All data were obtained from publicallyavailable sources and did not involve human participants.

Competing interestsThe authors declare that they have no competing interests.

Publisher’s NoteSpringer Nature remains neutral with regard to jurisdictional claims inpublished maps and institutional affiliations.

Author details1The Australian Prevention Partnership Centre, The Sax Institute, 10 JonesStreet, Ultimo, NSW 2007, Australia. 2School of Public Health and Social Work,Queensland University of Technology, Brisbane, QLD, Australia. 3PreventiveHealth Branch, Department of Health, Queensland Government, Brisbane,QLD, Australia. 4School of Public Health, Curtin University, Kent Street, GPOBox U1987, Perth 6845, Western Australia. 5Public Health Division,Department of Health, Government of Western Australia, 189 Royal Street,East Perth 6004, Western Australia.

Received: 30 May 2018 Accepted: 19 September 2018

References1. Institute for Health Metrics and Evaluation. Global Burden of Disease

Country Profile Australia. 2015; http://www.healthdata.org/sites/default/files/files/country_profiles/GBD/ihme_gbd_country_report_australia.pdf.Accessed 12 Nov 2017.

2. Australian Bureau of Statistics. Australian Health Survey: Consumption offood groups from the Australian dietary guidelines, 2011–12. 2016; http://www.abs.gov.au/ausstats/[email protected]/mf/4364.0.55.012 Accessed 12 Nov 2017.

3. National Health and Medical Research Council. Australian Dietary Guidelines- providing the scientific evidence for healthier Australian Diets. 2013;https://www.eatforhealth.gov.au/sites/default/files/content/n55_australian_dietary_guidelines.pdf. Accessed 9 Feb 2016.

4. Australian Bureau of Statistics. Australian Health Survey: Nutrition first results- foods and nutrients, 2011–12. 2014; http://www.abs.gov.au/AUSSTATS/[email protected]/DetailsPage/4364.0.55.0072011-12?OpenDocument. Accessed 12Nov 2017.

5. Australian Bureau of Statistics. Australian Health Survey: Updated results,2011–2012 - overweight and obesity. 2013; http://www.abs.gov.au/ausstats/[email protected]/lookup/33C64022ABB5ECD5CA257B8200179437?opendocument.Accessed 12 Nov 2017.

6. Lee A, et al. Monitoring the price and affordability of foods and dietsglobally. Obes Rev. 2013;14(Suppl 1):82–95.

7. Kettings C, Sinclair AJ, Voevodin M. A healthy diet consistent with Australianhealth recommendations is too expensive for welfare-dependent families.Aust N Z J Public Health. 2009;33(6):566–72.

8. Williams PG. Can the poor in Australia afford healthy food? Nutr Diet. 2011;68:6–7. https://doi.org/10.1111/j.1747-0080.2010.01485.x.

9. Ward PR, et al. Food stress in Adelaide: the relationship between lowincome and the affordability of healthy food. J Environ Public Health. 2013;2013:10.

10. Barosh L, et al. The cost of a healthy and sustainable diet – who can affordit? Aust N Z J Public Health. 2014;38(1):7–12.

11. Landrigan TJ, et al. Removing the Australian tax exemption on healthy foodadds food stress to families vulnerable to poor nutrition. Aust N Z J PublicHealth. 2017;41:591–7.

12. Vandevijvere S, et al. Monitoring and benchmarking population diet qualityglobally: a step-wise approach. Obes Rev. 2013;14(Suppl 1):135–49.

13. Lewis M, Lee A. Costing ‘healthy’ food baskets in Australia - a systematicreview of food price and affordability monitoring tools, protocols andmethods. Public Health Nutr. 2016;19(16):2872–86.

14. Carlson A, Frazao E. Are healthy foods really more expensive? It depends on howyou measure the price. In: Economic information bulletin no. (EIB-96): UnitedStates Department of Agriculture Economic Research Service; 2012. https://www.ers.usda.gov/webdocs/publications/44678/19980_eib96.pdf?v=42321

15. World Health Organization and Food and Agriculture Organization of theUnited Nations. Diet, nutrition and the prevention of chronic diseases, Reportof the joint WHO/FAO expert consultation. 2003; http://www.who.int/nutrition/publications/obesity/WHO_TRS_916/en/. Accessed 10 Nov 2017.

16. World Health Organization. Global Action Plan for the Prevention andControl of Non-communicable Diseases 2013–2020. 2013; http://www.who.int/nmh/publications/ncd-action-plan/en/. Accessed 10 Nov 2017.

17. Commission on Ending Childhood Obesity. Report of the Commission onEnding Childhood Obesity. 2016; http://www.who.int/end-childhood-obesity/publications/echo-report/en/. Accessed 10 Nov 2017.

18. Burns C, Sacks G, Gold L. Longitudinal study of consumer Price index (CPI)trends in core and non-core foods in Australia. Aust N Z J Public Health.2008;32(5):450–3.

19. Australian Bureau of Statistics. Consumer Price Index, Australia, Dec 2015Feature article: Australian dietary guidelines Price indexes. 2016; http://wwwabsgovau/AUSSTATS/abs@nsf/Previousproducts/64010Feature%20Article1Dec%202015?opendocument&tabname=Summary&prodno=64010&issue=Dec%202015&num=&view=. Accessed 12 Nov 2017.

20. Queensland Health. 2014 Healthy food access basket survey. 2015; https://wwwhealthqldgovau/research-reports/reports/public-health/food-nutrition/access/overview. Accessed 8 Feb 2016.

21. Pollard CM, et al. Geographic factors as determinants of food security: aWestern Australian food pricing and quality study. Asia Pac J Clin Nutr.2014;23(4):703–13.

22. Lee AJ, et al. Improving aboriginal and Torres Strait islander nutrition andhealth. Med J Aust. 2009;190(10):547–8.

23. Pollard C. Selecting interventions for food security in remote indigenouscommunities. In: Farmar-Bowers Q, Higgins V, Millar J, editors. Food securityin Australia: challenges and prospects for the future. Boston: Springer US;2013. p. 97–112.

24. Seal J. Monitoring the price and availability of healthy food - time for anational approach. Nutr Diet. 2004;61(4):197–8.

25. Burns C, Friel S. It's time to determine the cost of a healthy diet in Australia.Aust N Z J Public Health. 2007;31(4):363–5.

26. Hawkes C. Food taxes: what type of evidence is available to inform policydevelopment? Nutr Bull. 2012;37(1):51–6.

27. Sacks G, et al. ‘Traffic-light’ nutrition labelling and ‘junk-food’ tax: a modelledcomparison of cost-effectiveness for obesity prevention. Int J Obes. 2011;35(7):1001–9.

28. Veerman J, Cobiac L. Removing the GST exemption for fresh fruits andvegetables could cost lives. Med J Aust. 2013;199(8):534–5.

29. Lee AJ, et al. Testing the price and affordability of healthy and current(unhealthy) diets and the potential impacts of policy change in Australia.BMC Public Health. 2016;16:315.

30. Swinburn B, et al. INFORMAS (International Network for Food and Obesity/non-communicable diseases Research, Monitoring and Action Support):overview and key principles. Obes Rev. 2013;14(Suppl 1):1–12.

31. Australian Bureau of Statistics. Microdata: Australian health survey: nutritionand physical activity, 2011–12. 2013; http://www.abs.gov.au/ausstats/[email protected]/PrimaryMainFeatures/4324.0.55.002?OpenDocument. Accessed 12 Nov 2017.

32. Food Standards Australia New Zealand. AUSNUT 2011–2013. 2016; http://www.foodstandards.gov.au/science/monitoringnutrients/ausnut/Pages/default.aspx. Accessed 12 Nov 2017.

33. National Health and Medical Research Council. A Modelling System toInform the Revision of the Australian Guide To Healthy Eating. 2011;https://www.eatforhealth.gov.au/sites/default/files/files/the_guidelines/n55c_australian_dietary_guidelines_food_modelling_140121.pdf.Accessed 11 Feb 2016.

34. Xyris Software (Australia) Pty Ltd, FoodWorks Professional. 2012.35. National Health and Medical Research Council, Australian Government

Department of Health and Ageing, and New Zealand Ministry of Health.Nutrient Reference Values for Australia and New Zealand, IncludingRecommended Dietary Intakes 2006 Version 1.2 Updated September2017. 2006; https://www.nhmrc.gov.au/guidelines-publications/n35-n36-n37. Accessed 12 Nov 2017.

36. Australian Bureau of Statistics. Census of Population and Housing: Socio-Economic Indexes for Areas (SEIFA), Australia. 2011; http://www.abs.gov.au/websitedbs/censushome.nsf/home/seifa. Accessed 30 Oct 2015.

37. Australian Bureau of Statistics. Census of Population and Housing: Socio-economic indexes for areas (SEIFA), Australia: Mapping Files of StatisticalArea Level 2 Index of Relative Socio-economic Disadvantage, 2011. 2011;http://www.abs.gov.au/AUSSTATS/[email protected]/DetailsPage/2033.0.55.0012011?OpenDocument. Accessed 30 Oct 2015.

38. Australian Bureau of Statistics. Australian Statistical Geography Standard(ASGS): Volume 1 - Main structure and greater Capital City statistical areas,

Lee et al. Nutrition Journal (2018) 17:88 Page 13 of 14

Page 14: Healthy diets ASAP – Australian Standardised …...healthy ‘discretionary’ food choices, defined as foods and drinks high in saturated fat, added sugar, salt and/or alcohol that

July 2011. 2011; http://www.abs.gov.au/AUSSTATS/[email protected]/DetailsPage/1270.0.55.001July%202011?OpenDocument. Accessed 20 Oct 2015.

39. Google Maps. 2017; https://maps.google.com.au/. Accessed 14 Oct 2017.40. Microsoft Corporation, Microsoft Office 2007. 2007.41. Palermo C, et al. Development and evaluation of an iPad app for measuring

the cost of a nutritious diet. JMIR MHealth UHealth. 2014;2(4):e50.42. Australian Bureau of Statistics. Census 2011 Community profiles. 2011;

http://www.abs.gov.au/websitedbs/censushome.nsf/home/communityprofiles?opendocument&navpos=230. Accessed 1 Mar 2016.

43. Fair Work Ombudsman. Minimum Wages. https://www.fairwork.gov.au/how-we-will-help/templates-and-guides/fact-sheets/minimum-workplace-entitlements/minimum-wages. Accessed 22 Oct 2015.

44. Department of Human Services. Online Estimators. http://www.humanservices.gov.au/customer/enablers/online-estimators. Accessed 22 Oct 2015.

45. Australian Taxation Office. Simple Tax Calculator. www.ato.gov.au/calculators-and-tools/simple-tax-calculator/. Accessed 12 Nov 2017.

46. Australian Bureau of Statistics. Household Income and Wealth, Australia,2015–16. 2017; http://abs.gov.au/household-income. Accessed 12 Nov 2017.

47. Australian Government. Re:think; better tax, better Australia. Discussionpaper. 2015; http://bettertax.gov.au/publications/discussion-paper/.Accessed 15 Mar 2015.

48. Mhurchu CN, et al. Food prices and consumer demand: differences acrossincome levels and ethnic groups. PLoS One. 2013;8(10):e75934.

49. Lee J, Ralston R, Truby H. Influence of food cost on diet quality and riskfactors for chronic disease: a systematic review. Nutr Diet. 2011;68:248–61.

50. Rao M, et al. Do healthier foods and diet patterns cost more than lesshealthy options? A systematic review and meta-analysis. BMJ Open. 2013;3(12):e004277.

51. Friel S, Barosh LJ, Lawrence M. Towards healthy and sustainable foodconsumption: an Australian case study. Public Health Nutr. 2014;17(5):1156–66.

52. Jacobs, D., Perera, D., Williams, T. Inflation and the cost of living. ReserveBank of Australia Bulletin, March Quarter 33–46 2014; 2014. https://www.rba.gov.au/publications/bulletin/2014/mar/4.html. Accessed 12 Nov 2017.

53. Australian Council of Social Service. Minimum Wage Submission 2017. 2017;http://www.acoss.org.au/wp-content/uploads/2017/03/ACOSS-minimum-wage-submission-2017_FINAL.pdf. Accessed 10 Nov 2017.

Lee et al. Nutrition Journal (2018) 17:88 Page 14 of 14